Tumors are usually lethal to patients only when they grow beyond microscopic sizes. While it is evident that genetic mutations are the initiating steps in tumorigenesis, expansion of a lethal tumor mass is only achieved through increases in vascular supply. The process of new blood vessel growth from existing vessels or angiogenesis, is known to be regulated by both pro- and antiangiogenic proteins. One of the most widely studied pro-angiogenic regulators is the endothelial mitogen, Vascular Endothelial Growth Factor (VEGF). VEGF is produced primarily by tumor cells activating endothelial cells to promote blood vessel formation. However, an increase in angiogenesis also requires a concomitant decrease in negative angiogenesis regulators produced by both cancer and non-cancer cells and include endogenous inhibitors such as thrombospondin-1 and endostatin. Non-cancer cells that regulate tumorigenesis via controlling angiogenesis include both platelets and stromal cells. We have previously demonstrated that platelets store both pro- and antiangiogenic regulators within their alpha granules. Similarly it has been shown that non-tumor cells such as stromal cells or fibroblasts produce angiogenesis inhibitors such as thrombospondin-1. This proposal is based on the hypothesis that endogenous inhibitors of angiogenesis are critical regulators of tumor growth downstream of oncogenic activation and are mediated by platelet or stromal release. This hypothesis will be tested by: (1) determining if tumor growth and regression in mouse models of oncogene-driven cancers is angiogenesis-dependent;(2) determining how the platelet angiogenic proteome contributes to tumorigenesis;and (3) investigating how host expression of endogenous angiogenesis inhibitors regulate tumor growth. Summary: These studies may lead to novel drugs that can increase the body's own angiogenesis inhibitors. These experiments may also lead to the development of novel drugs that could direct platelets to release their collection of angiogenesis inhibitors into a tumor. From the planned studies may also emerge a simple blood test in which the angiogenesis regulatory molecules carried by platelets could be used to predict tumor recurrence long before symptoms or before visualization of the tumor becomes possible by conventional methods.